Type 2 diabetes mellitus affects a growing number of individuals worldwide, and U.S. racial minority groups in particular. States of excess cortisol secretion (Cushing's disease) are associated with hyperglycemia and diabetes mellitus. Mifepristone inhibits the action of glucocorticoids and is currently approved for the treatment of hyperglycemia in patients with Cushing's disease. We propose to study the efficacy of mifepristone in patients with usual type 2 diabetes mellitus not associated with Cushing's disease, wherein more subtle abnormalities of cortisol action may still play a role in adversely affecting glucose control. Because mifepristone may also have gynecological side effects due to anti-?progesterone effects and is also used clinically by women for pregnancy termination, we will restrict this initial pilot study to men only; if this project determines that it has favorabl actions on glucose lowering in men, it would then justify replicating this study in women to assess its glycemic benefits against those potential gynecological side effects. This pilot project will enroll 60 predominantly racial minority male subjects with type 2 diabetes mellitus inadequately controlled on combination oral agents. They will be randomized to receive either mifepristone 600 mg daily or matching placebo tablets, in a double blind fashion, for 3 months. All concurrent medication dosages will be kept constant, and consistent dietary and lifestyle counseling will be provided. Changes in glucose control, body weight and body composition, measures of insulin sensitivity and beta cell function from oral glucose tolerance testing, lipids, blood pressure, and clinical safety will be compared between groups. At baseline and at 3 months, a subset of 10 subjects will also undergo a euglycemic hyperinsulinemic clamp procedure to determine physiological responses to a standardized intravenous infusion of high-?dose insulin, measuring insulin-?mediated glucose uptke, insulin mediated suppression of hepatic glucose output, and suppression of non-? esterified fatty acids. Indirect calorimetry will also be performed simultaneously to determine changes in rates of insulin-?stimulated glucose oxidation, non-?oxidative glucose disposal and suppression of lipid oxidation. If successful, our findings will 1) support further follow-?up studies to delineate a potential clinical role for mifepristone as a new pharmaceutical choice for the treatment of common-?variety type 2 diabetes; 2) support further molecular, cellular and in vivo investigations into the mechanisms of how usual physiological levels of glucocorticoids contribute to the pathophysiology of type 2 diabetes, both at the intracellular and whole organism levels; and 3) stimulate greater research productivity for the PI towards achieving independent funding for additional studies aimed at better understanding all of the above research aims.